Method and system to achieve heating and cooling temperature within one or more zones when occupied
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Building managers face challenges in determining the optimal time to activate HVAC systems for heating or cooling enclosed spaces before occupancy, leading to inefficient operation and increased costs due to prolonged system activation.
Innovation Solution
A method and system utilizing a machine learning model trained on historical temperature and input conditions to determine a threshold time period for achieving desired temperature set points in occupied zones, adjusting temperature set points using AI/ML techniques to optimize HVAC operation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the HVAC system is activated for a longer period before occupancy, then the desired temperature is achieved more reliably, but the operating cost increases
Solution Approach 1:
The system performs preliminary action by activating the HVAC system in advance of occupancy, but optimizes the timing using machine learning predictions. The ML model determines the precise threshold time period needed before occupancy to achieve the desired temperature, avoiding both premature shutdown and excessive pre-heating/cooling. This resolves the contradiction by finding the optimal preliminary action duration that ensures temperature reliability while minimizing energy waste.
2Ease of operation
If the HVAC system is activated earlier to ensure temperature comfort, then occupant comfort is improved, but energy consumption increases
Solution Approach 1:
The system applies dynamics by transitioning from static, fixed pre-heating/cooling schedules to dynamic, adaptive timing based on real-time predictions. The machine learning model continuously learns from historical data and adjusts the threshold time period dynamically according to varying conditions such as weather, occupancy patterns, and zone characteristics. This ensures occupant comfort is maintained while optimizing energy consumption by avoiding unnecessary early activation.
3Device complexity
If manual prediction methods are used to determine activation time, then system complexity is reduced, but prediction accuracy becomes impossible
Solution Approach 1:
The system introduces an intermediary - the machine learning model - that acts as a bridge between simple system architecture and accurate prediction requirements. The ML model processes multiple input parameters (historical temperature data, weather conditions, occupancy patterns, zone characteristics) and outputs precise threshold time period predictions. This intermediary enables high prediction accuracy without requiring complex manual calculations or overly complicated system architecture, as the ML model encapsulates the complexity in a manageable computational layer.
Data Source
AI summary
A method and system to achieve heating and cooling temperature within one or more zones when occupied is disclosed. The method comprises training, via at least one processor, a machine learning (ML) model based at least on historical temperature data and input conditions of one or more zones; receiving temperature data from one or more sensors and one or more input conditions of the one or more zones for a predefined time period in real-time; determining a threshold time period to achieve heating or cooling temperature within the one or more zones when occupied using the trained ML model, based at least on the received temperature data and input conditions of the one or more zones for the predefined time period; and adjusting the one or more temperature set points to achieve heating or cooling temperature within the one or more zones when occupied.